This invention relates to groups of humans or groups of artificially intelligent machines, specifically to improving their efficiency in attaining group goals.
As we enter the twenty-first century increasingly more of the problems that must be solved by humans are complex in nature. Their solutions require the cooperative effort of group members with different but complementary skills and experience who must interact quickly and efficiently to craft a solution. Groups of humans, when functioning as intended are commonly characterized as teams. Webster defines a xe2x80x98teamxe2x80x99 as xe2x80x9ca number of people working together on a common taskxe2x80x9d1. The literature refines this definition as follows: xe2x80x9cA team is a small number of people with complementary skills who are committed to a common purpose, performance goals, and approach for which they hold themselves mutually accountablexe2x80x9d2.
1Webster""s Encyclopedic Dictionary of the English Language, 1992; Lexicon Publications. 
2The Wisdom of Teams; creating the high-performance organization; Jon R. Katzenbach, Douglas K. Smith; Harvard Business School Press, 1993 Boston Mass. 
Although reliance by humans on group effort predates recorded history, a high performance team effort is a rare occurrence. This is because as the literature states, xe2x80x9cIndividuality and self preservation remain the rule, shared responsibility based on trusting others is the exception. A reluctance to take a risk and submit one""s fate to the performance of a team, therefore, is almost inbredxe2x80x9d3. When high performance teams do occur they have the following characteristics:
xe2x80x9cParticipative Leadershipxe2x80x94create inter-dependency by empowering, serving others;
Shared Leadershipxe2x80x94all members feel responsible for performance of work unit;
Aligned on Purposexe2x80x94common purpose;
High Communicationxe2x80x94trust, open honest communication;
Future Focusedxe2x80x94change is opportunity;
Focused on Taskxe2x80x94focus on results;
Creative Talentsxe2x80x94applying individual talents and creativity;
Rapid Responsexe2x80x94identifying and acting on opportunity.xe2x80x9d4 
3The Wisdom of Teams; creating the high-performance organization; Jon R. Katzenbach, Douglas K. Smith; Harvard Business School Press, 1993 Boston, Mass. 
4Creating the high-performance team; Steve Buchholz, Thomas Roth; Wiley, 1987. 
Maximizing a team""s efficiency requires that all members"" actions add to team synergy. Given the natural reluctance to team orientation, only a supportive value/reward system can encourage team-oriented action. An old business adage comes into play, xe2x80x9cyou get what you xe2x80x98inspectxe2x80x99 not what you xe2x80x98expectxe2x80x99xe2x80x9d. By this axiom a team member will be motivated to maximize his or her performance as measured, because the value/reward system is based on what is measured. When the team-member measurement criteria is not directly aligned with team goals, the value/reward system motivates team members to act in ways that are counter-productive to a team winning effort.
Team value/reward methods have not changed significantly since pre-historic times. There are two basic team value/reward methods. Either team members are valued for team membership, or team members are valued for their individual skills. In the first case, the hunting party that killed the saber-toothed tiger all shared its meat. In the second case, individual members of the hunting party who created the weapons, possessed the cunning to trap the prey, and demonstrated the bravery, strength and foot speed to accurately plunged the spear into the beast, were rewarded individually for their valued skills. However, neither of these methods, separately or in combination, measures or instills all of the high performance team characteristics listed above.
The deficiencies of these methods are as follows:
a) Provide no means to establish a direct cause/effect relationship between individual team-member actions or collaborations, and team goal achievement:
either, they value and measure team-members"" individual accomplishments regardless of their effect on team achievement, which re-enforces the individual""s self-centered focus instead of a focus on team results, i.e. reward excellence in spear-making and spear-throwing;
or, they value all team members for team achievement regardless of their contribution to team results, i.e. all share the saber-tooth tiger meat;
b) Provide no means to identify and expose for analysis and reward, the critical few team-member actions and collaborations that are causal to team achievement; provide no direction to team members on how to apply individual talents and creativity to the achievement of a common task;
c) Provide no means to establish a common understanding of the value of different roles played by team members and thereby enable all team members to be accountable for team performance; provide no mechanism that directly relates the value of different functional types of team-member actions, to the achievement of a team goal, i.e., provide no means to compare the relative importance of making the deadly weapon, to creating the deadly trap, or to killing the beast;
d) Provide no means for a group to integrate its activities in pursuit of a group goal, when given only partial knowledge of how to achieve it; furthermore, provides no means for valuing and measuring collaborations between two or more team members, that directly result in team goal achievement i.e. provide no means to determine the spear-thrower/trap-layer collaborations which have the best chance of achieving a kill;
e) Provide no means to encourage trust or open, honest communications to enable a team to rapidly identify and act on opportunities; i.e. provide no protocol for interaction between spear-throwers and trap-layers to enable them to rapidly identify and act on opportunities and thereby increase the chance of a kill; provide no method or mechanism to fairly and objectively measure and record team-member activities;
This invention defines a method and performance feedback system to enable group members to interact quickly and efficiently with their peers to craft a solution to a group goal without external intervention. Using the common understanding provided by this performance feedback system, team members effectively collaborate to achieve a common goal while possessing only partial understanding of how to achieve it. This method and system enables the group to progress towards its common goal by incorporating internal creativity and group intuition as modes of behavior, rather than internal conflict and internal competition.
The present invention achieves the following desirable features:
a) means to provide understanding of individual team-members"" performance by exposing the direct cause/effect relationship between individual team-member actions and team goal achievement; thus enabling team members to adjust their actions and inter-actions with other team members, to enhance future team performance; furthermore, the values and measurements produced using this method and system can be used to gratify the team-members"" individual needs in direct proportion to their measured value to team achievement; furthermore, this fair reward and recognition for effective team activity works to reduce team-members"" conflict between the natural tendency toward individuality and self-preservation, and the counter-intuitive human contrivance of team orientation;
b) means to expose for analysis and reward, the critical few team-member actions and collaborations that are causal to team achievement; furthermore, this invention only values individual team-member actions that are part of an action sequence that directly results in the achievement of a valued team goal; examples from past successes enable team members to learn how to apply individual talents and creativity to improve future team performance;
c) means to compare the values of different functional roles team-members play when contributing to team achievement, and to enable all team members to be accountable for team performance; a method and system that directly relate the value of different functional types of team-member actions, to the achievement of a team goal;
d) means to guide team members on how to integrate their activities in pursuit of a group goal, when given only partial understanding of how they can achieve it; furthermore, a means for valuing and measuring collaborations between two or more team members, that directly result in team goal achievement; furthermore, a common team-member interaction protocol that stimulates the development of quick and efficient teammate inter-creativity and group intuition;
e) trusted performance feedback method and system to promote open, honest communication among team members and thereby enable them to rapidly identify and act on opportunities; this method and system fairly and objectively measure and record valued team-members actions and collaborations and sums, compares, stores and publishes the values of these actions and collaborations.
These improvements are accomplished by the present invention by providing the following abstract objects, which are shown in FIG. 1:
1. process flow abstraction called COMPETITIVE ENVIRONMENT that represents a domain with a set of rules and conditions that determines the team-member actions which are Contested and which are Uncontested, and determines the relative value of functionally different team-member actions, thereby influencing the activity which team-members employ while contesting to achieve a team goal.
2. process flow abstraction called CONTEST that represents a discrete event within a COMPETITIVE ENVIRONMENT during which one or more teams strive to achieve a common goal within a predetermined amount of time.
3. process flow abstraction called POSSESSION that represents a unique pursuit of a team goal, beginning with the successful acquisition of the resources required to achieve that goal, and proceeding with the enhancement of those resources by stages until the team goal is achieved, the resources are lost, or the CONTEST duration is exceeded.
4. process flow abstraction called TEAM ACHIEVEMENT VALUE that represents the result of a team""s pursuit of its goal during a POSSESSION or CONTEST, and having the value 1=achievement (win), or 0=non-achievement (loss).
5. process flow abstraction shown in FIG. 2, called ACT SEQUENCE that represents the critical few team-member actions and their sequence, that are causal to TEAM ACHIEVEMENT VALUE=1. The ACT SEQUENCE is comprised of three discrete sequential stages; first an ACQUIRING ACTION that acquires control of resources prerequisite to attain TEAM ACHIEVEMENT VALUE=1; followed by a CREATING ACTION that removes variable obstacles creating an uncontested TESTING ACTION; followed by a TESTING ACTION that eliminates the fixed obstacles to attain TEAM ACHIEVEMENT VALUE=1. An ACT SEQUENCE with one or more completed stages is called a POSSESSION. The ACT SEQUENCE, or POSSESSION is successful if it results in TEAM ACHIEVEMENT VALUE=1. Moving directly through the ACT SEQUENCE maximizes team efficiency. Therefore, to focus team members"" efforts on team achievement rather than on performing isolated actions, this invention stipulates that only the last A, C and T Stage actions or collaborations are valued for every successful ACT SEQUENCE.
6. process flow abstraction called JUDGE, representing an expert human or artificially intelligent decision-making entity, that monitors a CONTEST and identifies and selects valued team-member actions and collaborations within the context of the TEAM-MEMBER INTERACTION PROTOCOL (TIP) shown in FIG. 3.
7. process flow abstraction called TEAM-MEMBER INTERACTION PROTOCOL that governs team-member interaction during a CONTEST. The protocol stipulates low priority two-way peer-to-peer communication among all PLAYERS and overlays a higher priority star topology two-way communication which places the empowered PLAYER with the role as STEWARD at the center of the star. The Steward""s communications have highest priority.
8. process flow abstraction within the TEAM-MEMBER INTERACTION PROTOCOL called PLAYER that represents a team member who performs actions and interacts with other team members in pursuit of a team goal.
9. process flow abstraction within the TEAM-MEMBER INTERACTION PROTOCOL called STEWARD which represents a role taken on by a PLAYER that empowers that PLAYER to protect team resources and advance a POSSESSION to the next stage in the ACT Sequence.
10. unit of measure called TEAMSHARE that represents one team member""s share of the reward for a successful POSSESSION or CONTEST. TEAMSHARE is calculated as the aggregate of an individual team member""s ACQUIRING ACTIONs, CREATING ACTIONs and TESTING ACTIONs that are part of successful ACT SEQUENCEs. The aggregate of all team-members"" TEAMSHAREs is equal to the total reward value attributed to a team. TEAMSHARE serves as the common currency for comparison of different team members"" contributions to TEAM ACHIEVEMENT VALUE=1.
In addition, the present invention provides the following mechanisms also shown in FIG. 4 and FIG. 14:
1. A mechanism called Acton-Type Valuing Mechanism 100 for determining the statistical correlation of functionally different team-member action-types, to TEAM ACHIEVEMENT VALUE. This mechanism is comprised of a method and collection of co-operating software programs executing on one computer or plural computers on a network, that process a sample of CONTEST data to produce a table of valued team-member action-types with associated value coefficients for a particular COMPETITIVE ENVIRONMENT.
2. A mechanism called Action Scoring Mechanism 300 for determining and assigning a POSSESSION""s reward value to the contributing team members for their valued actions. This mechanism is comprised of a method and collection of co-operating software programs executing on one computer or plural computers on a network, that use POSSESSION data received from a Action Selection Mechanism, and Action-Type/Value Coefficient data from the Acton-Type Valuing Mechanism as inputs, to produce as output a collection of team members with their associated Teamshares.
3. A mechanism called Action Selection Mechanism 200 for identifying, selecting, collecting and validating pertinent CONTEST data, and utilizing the services of an expert JUDGE to monitor the CONTEST. JUDGEs use the context of the TEAM-MEMBER INTERACTION PROTOCOL to help them focus attention on high potential team-member activity, enabling them to rapidly identify valued team-member actions and collaborations. This mechanism is comprised of a method and collection of software programs executing on one computer or plural computers on a network, and one or more JUDGEs who input the data.
4. A Performance Feedback Mechanism shown in FIG. 4, that collects CONTEST data and controls and interacts with the Acton-Type Valuing Mechanism, Action Scoring Mechanism, Action Selection Mechanism, database and Internet server to calculate team-members"" TEAM SHAREs, and stores and publishes the resulting information.
5. Observer Mechanism shown in FIG. 14 that performs the JUDGE role, for a team of artificially intelligent machines or computer PLAYERs. This mechanism consists of neural network software programs called Observers that monitor the communication patterns and data-transfers between PLAYERs in the context of the TIP. This mechanism, through its self-learning neural networks, identifies potentially valuable sequences of activity for scrutiny by the Action Selection Mechanism without requiring all alternatives to be explored. This mechanized judging function enables a group of artificially intelligent machines or computers to self-manage their interactions while they craft a solution to a group goal given only partial knowledge of how to achieve it.
Those and other improvements are set forth in the following detailed description. For a better understanding of the present invention with advantages and features, refer to the description and to the drawings.
Basic Structure of a Preferred Embodiment of the Invention In a preferred embodiment of the present invention, the team-member Performance Feedback System 500 is a collection of co-operating computer programs executing on one computer or plural computers connected by a network, as shown in FIG. 4. This particular implementation of the invention utilizes standard Microsoft Windows application programs called SPSS Base 8.0, SPSS Professional Statistics 8.0 and SPSS Advanced Statistics 8.0 and IBM DB2 database to perform standard statistical calculations and to store information on the computer. The computer programs that implement work flow are implemented as Java objects in the form of Java Servlets, Java Beans, Enterprise Java Beans and Java Server Pages. Statistical calculations are performed using SPSS Corporation""s Correlation Analysis and Logistic Regression Analysis programs. Standard Internet protocols, languages and browsers are employed to implement the user interface. However, any general-purpose message passing mechanism and programming language is sufficient to implement the features of the present invention. For the purposes of teaching this invention, the following description of a preferred embodiment uses the general concepts of messages and entities, instead of object-oriented terminology.
Action-Type Valuing Mechanism 100
The Action-Type Valuing Mechanism shown in FIG. 5 establishes which team-member action-types are valued. It also classifies each valued action-type by ACT Sequence Stage and then calculates the relative value of each of these action-types by creating a Value Coefficient for each ACT Sequence Stage. The Action-Type/Value Coefficient relation produced by this mechanism is a required part of the definition of a COMPETITIVE ENVIRONMENT and must be established prior to the operation of the team-member Performance Feedback System.
Step 1: Action-Type Definition 110
A statistically significant sample of CONTEST data is input into the Action-Type Valuing Mechanism. This input data includes the number of occurrences of each Action-Type and the value of TEAM ACHIEVEMENT VALUE (1 or 0) for the CONTEST. Action-Type Definition is accomplished by subjecting these CONTEST data to SPSS Corporation""s Correlation Analysis Program. Action-Types with a significant correlation to TEAM ACHIEVEMENT VALUE=1 are selected as Valued Action-Types. The output of the Action Definition Step is the Action-Type Value entity with the following attributes added: COMPETITIVE ENVIRONMENT, Qualified Action-Type.
Step 2: Classification of Action-type by ACT Sequence Stage 120
Qualified Action-Types from Step 1 are classified by ACT Sequence Stage according to the following criteria:
1. ACQUIRING ACTION acquires resources necessary for TEAM ACHIEVEMENT VALUE=1
2. CREATING ACTION removes variable obstacles, thus enabling an uncontested TESTING ACTION.
3. TESTING ACTION eliminates the fixed obstacles to attain TEAM ACHIEVEMENT VALUE=1
The output of Step 2 is the Action-Type Value entity with the following attribute added: ACT Sequence Stage.
Step 3: ACT Sequence Stage Efficiency Definition 130
Qualified Action-Types for the COMPETITIVE ENVIRONMENT from step 2 are inserted in the following general equations as source variables:
1. ACQUIRING ACTION EFFICIENCY for a CONTEST equals the sum of My Team""s Contested ACQUIRING ACTIONs divided by the sum of the CONTEST""s Total ACQUIRING ACTION opportunities,
2. CREATING ACTION EFFICIENCY for a CONTEST equals the sum of My Team""s Uncontested TESTING ACTIONs divided by the sum of My Team""s Total POSSESSIONs,
3. TESTING ACTION EFFICIENCY for a CONTEST equals the sum of the Values of My Team""s TESTING ACTIONs divided by the sum of the Potential Values of My Team""s Testing Action Opportunities.
These ACT Sequence Stage Efficiency equations may be optionally tested using the sample CONTEST data input in step 1, using SPSS Corporation""s Correlation Analysis Program. This test insures that the ACT Sequence Stage Efficiencies have a low correlation with each other and a low correlation with any of their source variables. Low correlation indicates that the ACT Sequence Stage Efficiencies are independent of one another and not merely a composite of the source variables, thereby qualifying the ACT Sequence Stage Efficiencies as independent variables for a mathematical model of TEAM ACHIEVEMENT VALUE.
The outputs of the ACT Efficiency Definition Step 3 are definitions for ACQUIRING ACTION EFFICIENCY, CREATING ACTION EFFICIENCY and TESTING ACTION EFFICIENCY in terms of the Qualified Action-Types for the COMPETITIVE ENVIRONMENT specified in Steps 1 and 2.
Step 4: ACT Sequence Stage Valuing 140
Action-types classified in different ACT Sequence Stages may have different impacts on TEAM ACHIEVEMENT VALUE. This impact is determined as follows. A Logistic Regression Analysis software program is used to calculate a Logistic Regression Model with TEAM ACHIEVEMENT VALUE as the dichotomous (1 or 0) dependent variable and the ACT Sequence Stage Efficiencies as independent variables. Inputs for this Step are the sample CONTEST data input into Step 1, the Action-Types and ACT Sequence Stage outputs of Step 2, and the ACT Sequence Stage Efficiency definitions from step 3. Step 4 produces a mathematical model of the cause/effect relationship between team-member actions and TEAM ACHIEVEMENT VALUE including a constant coefficient associated with each ACT Sequence Stage Efficiency term. The resulting mathematical model equation follows:
TEAM ACHIEVEMENT VALUE equals the constant a, plus the Value Coefficient b times the ACQUIRING ACTION EFFICIENCY, plus the Value Coefficient c times the CREATING ACTION EFFICIENCY, plus the Value Coefficient d times the TESTING ACTION EFFICIENCY.
TEAM ACHHIEVEMENT VALUE can have values 1=achievement, 0=non-achievement. The three Value Coefficients, b, c, d indicate the value of an action-type from one ACT Sequence Stage relative to the value of an action-type from one of the other two ACT Sequence Stages in the same COMPETITIVE ENVIRONMENT. The output of Step 4 is the Action-Type Value entity with the following attribute added: Value Coefficient
Action Selection Mechanism 200
The Action Selection Mechanism shown in FIG. 6, functions to reveal the critical few team-member actions and collaborations that directly contribute to TEAM ACHIEVEMENT VALUE=1, from the complete set of actions that comprise a CONTEST. This invention requires judgement and experience to select and record all appropriate team-member actions for assignment of value. JUDGES perform this function. The effectiveness of this invention rests on the consistent, objective interpretation of team activities by the JUDGES. During a CONTEST many actions are occurring simultaneously. It is the responsibility of the JUDGE to focus on the potentially valuable action sequences so that all action sequences do not have to be scrutinized. The JUDGE, whether human or artificially intelligent software program, utilizes the TEAM-MEMBER INTERACTION PROTOCOL (TIP) as an aid to focus its attention on potentially valuable action sequences. The TIP is also embodied in the Action Selection Validation software program to validate the required flow of the ACT Sequence shown in FIG. 2 and the validation of selected collaborative actions. From these sequences, the JUDGE selects valued actions in compliance with the following validation rules that are implemented in the Data Validation software program 220 executing on a computer used to collect CONTEST data:
1. Only team-member actions, whose action-type is defined in the output of Step 1 of the Action-Type Valuing Mechanism, are valued.
2. Only the final action or collaboration in each stage of an ACT Sequence is valued. The flow of action must move sequentially forward from ACQUIRING ACTION to CREATING ACTION to TESTING ACTION. If the flow takes a step back before moving forward again, the most recent action or collaboration in an ACT Sequence Stage is valued and the prior action or collaboration is discarded.
3. An ACT Sequence requires only a TESTING ACTION. In this case, the TESTING ACTION accrues the value for the CREATING ACTION as well.
4. ACQUIRING ACTIONs that are uncontested are not valued. TESTING ACTIONS that result from testing opportunities awarded due to violations of the rules of the COMPETITIEVE ENVIRONMENT are valued.
ACT Sequence selections made by the JUDGE are validated by the Action Selection Mechanism prior to the passing of data to the Action Scoring Mechanism. The number of JUDGES required varies by COMPETITIVE ENVIRONMENT and depends on the level of activity detail that must be scrutinized. To aid a human JUDGE in collecting CONTEST data while simultaneously judging the CONTEST, the Action Selection Mechanism includes a data entry software program that enables the human JUDGE to select from menus of pre-programmed options instead of entering raw data while judging the CONTEST. The output of the Action Selection Mechanism is a POSSESSION entity with the following attributes added:
ACT Sequence, ACT Sequence Stage, Actions, Team Members
The Action Selection Mechanism optionally includes the following Data Collection Apparatus shown in FIG. 7, to enable human JUDGES to accurately capture data in xe2x80x9creal timexe2x80x9d situations that require the observation of fast moving team activity. These machine functions facilitate capture and validation of data:
1) Voice recognition software program and head-mounted microphone 410 are used by the JUDGE to facilitate data collection. They enable predefined audible commands to be recognized as computer commands during data capture. A pre-programmed set of audible selections enable the JUDGE to capture CONTEST data by voice commands.
2) A miniature head-mounted video screen 420 in the JUDGE""s field of view, displays the current state of the data collection process, and enables her to see a current menu of computer command selections from which she can audibly select computer commands by using the voice recognition apparatus described above. This enables the JUDGE to validate that her computer commands are being processed as intended without significantly diverting her attention away from observing the CONTEST.
Action Scoring Mechanism 300
The Action Scoring Mechanism shown in FIG. 8 distributes the reward value-of an ACT Sequence among the contributing team-members.
Step 1: Determine Active ACT Sequence Stages 310
ACT Sequence data from the Action Selection Mechanism is used as input to the Action Scoring Mechanism. The ACT Sequence data is scanned to determine which of its ACT Sequence Stage values are not null, meaning that they contributed to the ACT Sequence""s success. If the ACQUIRNG ACTION stage value is not null, it is Active, meaning that an ACQUIRING ACTION contributed to the ACT Sequence""s success. Consequently, it qualifies as a contested action and should receive a share of the ACT Sequence reward. The ACT Sequence data is then scanned to determine if the CREATING ACTION stage is not null, indicating that it is Active and qualifies as a contested action and should receive a share of the ACT Sequence reward. If the CREATING ACTION is null then the team-member performing the TESTING ACTION receives the CREATING ACTION""s share of the reward also since the TESTING ACTION was accomplished without the benefit of being uncontested. The output of this stage is an Action Scoring Value entity with the following attributes added: ACT Sequence Stage, Stage Active (true or false).
Step 2: Reward Active ACT Sequence Stages 320
Each of the Active Stages in the ACT Sequence shares the ACT Sequence""s reward. Input to Step 2 consists of the output from Step 1, and the ACT Sequence Reward included in the input to Step 1 and the output of the Action Valuing Mechanism. The ACT Sequence Stage Sharing Factor for an Active Stage is calculated by the following equations using the Value Coefficients obtained from the Action-Type Valuing Mechanism:
1. Acquiring Stage Sharing Factor equals Value Coefficient b divided by the sum of Value Coefficients b, and c, and d.
2. Creating Stage Sharing Factor equals Value Coefficient c divided by the sum of Value Coefficients b, and c, and d.
3. Testing Stage Sharing Factor equals Value Coefficient d divided by the sum of Value Coefficients b, and c, and d.
If a Stage is inactive, its corresponding Value Coefficient is zero in the above Sharing Factor equations. The reward for each Active Stage is calculated by multiplying the ACT Sequence Reward value by the Sharing Factor for the Active Stage. The output of Step 2 is the Action Scoring Value entity with the following attribute added: ACT Sequence Stage Reward
Step 3: Identify Contributing Team Members 330
The reward sharing process continues by identifying the one or more team members who performed the action or collaboration that contributed to each Active Stage. They will share the Stage Reward equally. Input to Step 3 is the output from Step 2 and the input to Step 1. The output of Step 3 is an Action Scoring Value entity with the following attributes added: Team-Members, Action-Types.
Step 4: Share Active ACT Sequence Stage Reward 430
The Stage Reward is shared with the one or more team members who performed the actions and collaborations that contributed to each Active Stage. Input to Step 4 is the output of Step 3. When collaboration occurs, the Stage Reward is divided equally among the actions associated with that Stage. The resulting TEAMSHARE value is associated with each action. The output of Step 4 is an Action Scoring Value entity with the following attribute added: TEAMSHARE.
Performance Feedback Mechanism 510
The Performance Feedback Mechanism (PFM) implements the ACT Sequence Stage Valuing and Performance Feedback functionality of this invention. As shown in FIG. 10, the PFM acts as a controller for the Performance Feedback System (PFS) by controlling the interaction of the Action-Type Valuing Mechanism 100, and Action Selection Mechanism 200, and Action Scoring Mechanism 300, and the database and the Internet Server. Prior to the operation of the PFS, CONTEST data is input into the data collection software program in the Action Selection Mechanism 200. There it is formatted and validated and then passed to the Action-Type Valuing Mechanism 100 which adds ACT Sequence Stage Value Coefficients that are needed to complete the definition of the COMPETITIVE ENVIRONMENT. Once the PFS is operational, CONTEST data is input into the data collection software in the Action Selection Mechanism 200 where it is validated and composed into a POSSESSION entity and passed to the Action Scoring Mechanism 300 where the TEAMSHARE values are assigned. The Performance Feedback Mechanism calculates the TEAMSHARE on an aggregate basis and stores and retrieves it from the database.
The Performance Feedback Mechanism (PFM) 510 is implemented as a Java Servlet that executes on the Application Server computer shown in FIG. 9. It receives commands from the user interface via the Internet browser. User commands are input to the PFM via the Internet server. The Internet server forwards these commands using standard Internet protocols and languages. The PFM supplies the data and instructions to produce Java Server Pages, which are sent to the Internet server to respond to user requests. Web page caching may be used as an option to speed page delivery to the user.